Introduction

Cardiac Magnetic Resonance (CMR) provides a comprehensive non-invasive characterization of acute myocardial infarction (MI).

Purpose

to test whether a single CMR study may improve the prognostic stratification of MI patients.

Methods

29 patients were studied on day 3 ± 1 after a first MI, between June 2006 and April 2007. CMR included SSFP and T2 STIR sequences, first pass perfusion, Early (EGE: 1 to 2 minutes) and Late (LGE: 10 to 15 minutes) Gadolinium Enhancement. Subsequent clinical management was not affected from CMR results, and patients underwent the usual clinical and echocardiographic follow-up. In 2009 we tested the occurrence of the end-point of cardiac death or left ventricular (LV) remodelling (EDVi > 90 ml/m2 or LVEF < 50% at echocardiographic follow-up) in the studied population.

Results

At 23 ± 5 months follow-up, the endpoint was observed in 7 pts (25%): 1 cardiac death, 6 adverse LV remodeling. See Table 1 for the prognostic role of different CMR parameters: among the tested parameters, myocardial haemorrhage, observed in 8 pts (27%), has a unique 96,5% accuracy in predicting the endpoint, with a 100% NPV. Unexpectedly, LGE-based parameters such as total infarct mass and the presence MVO show prognostic accuracy similar to the echocardiographic LVEF <50%. The 7 pts reaching the endpoint show both MVO (globally present in 15 pts -51%-) and Myocardial Haemorrhage. It seems that the assessment of MVO should keep in account the total amount and not only the mere presence of MVO: this hampers the usefulness of MVO, since the total amount changes over time.

Table 1
Full size table

Conclusion

T2-STIR images allow a long-term prognostic stratification of unrivalled accuracy during the acute phase of MI. The use of gadolinium do not significantly improve the prognostic accuracy of CMR.